Excavation

Date August 1988 - September 1988

Event ID 1005812

Category Recording

Type Excavation

Permalink http://canmore.org.uk/event/1005812

Rescue excavation of four areas within this moated site was carried out in advance of serious plough erosion. The scarped natural platform is sub-rectangular in plan and no more than 3m in height. The defences were found to consist of three concentric rings of ditches with two ploughed-out counterscarp banks or ramparts.

Area I ran up the slope from the inner ditch to the N summit lip. The ditch was steeply V-cut through granite, 7m wide by 2.4m deep. This was backed to S by the inner rampart base which was originally 4.5m wide. A sub-rectangular stone lined pit was found against the rampart back.

This measured 1.7m wide by 1.4m deep; it had been a storage pit, backfilled with midden material containing c 120 sherds of medieval pottery including fragments of a knight's jug. S of the pit was a 3m square area of large worn cobbles including a recessed drain. This was seen as the surviving part of a late medieval courtyard which may have covered most of the interior. This small area alone survived as it had not been ploughed.

Area II was in the middle of the platform, where later stone buildings were anticipated. All that had survived the plough were the bases of deeply cut post holes, which had formed part of at least two large timber buildings. Approximately 0.30m has been lost off the crown of the moated platform. Area III was located on the SW side of the mound, and provided a 30m long transect through the multiple defensive lines and into the castle interior. The 5m wide outer ditch here appeared to be secondary as it was cut through the collapsed turf facing of the mid rampart. This bank base was 7.6m wide, and late on in the sequence had been cut by the narrow mid ditch (possible palisade trench). The inner ditch was 8.9m wide by 2.1m deep with a V-shaped profile. Slight evidence of a recut was found on the inner side. A two phase rampart base, 8.6m wide, was found inside the inner ditch. Both ramparts sealed old ground surfaces with a maximum thickness of 0.20m, and these exhibited signs of extensive prehistoric cultivation and/or occupation. Features inside the inner rampart were well preserved, and a rampart terminal was revealed containing stone and timber structural features which may have been part of an entrance tower. No trace of a later medieval curtain wall was found, and it seems likely that the earthwork defences were maintained into the 17th century when the castle was finally abandoned.

Area IV was located at the base of the naturally well defended E side of the mound. Waterlogging was anticipated here and the sole aim was the recovery of environmental samples. A 2m thick layer of peat was exposed 1m below ground level, sealing a very deep layer of waterlogged sandy soil. Individual fragments of wood were recovered along with a sample of what may be matted pine needles, resembling a forest floor. No evidence of the documented chapel was found, although a discovery in a nearby dyke raised the likelihood of its existence. The top part of a 17th century tombstone was discovered, which lacking any other nearby source could only have come from the castle chapel, which may have ended its life as a family burial ground.

The bridge at NJ 5918 2889 may be 17th century in date and reflects the location of the ancient approach road to the castle. An experimental Subsurface interface Radar survey was carried out with sponsorship from Oceanfix Ltd, Aberdeen. This technique provided a series of vertical slices through the NW part of the site, and displayed detailed information of buried structures, ditch cuts and individual fills.

P Yeoman 1988.